SEISMIC BEHAVIOR OF SILOS WITH DIFFERENT HEIGHT TO

DIAMETER RATIOS CONSIDERING GRANULAR MATERIALSTRUCTURE INTERACTION

F. Nateghi and M. Yakhchalian*
Structural Engineering Research Center, International Institute of Earthquake Engineering and Seismology,
Tehran, Iran
nateghi@iiees.ac.ir, m.yakhchalian@iiees.ac.ir
*Corresponding Author
(Received: April 09, 2011 Accepted in Revised Form: December 15, 2011)

doi: 10.5829/idosi.ije.2012.25.01b.04

D
I

Abstract Silos are structures that are used for storing different types of granular material. Dynamic
behavior of silos under seismic loads is very complex. In this paper seismic behavior of steel silos with
different height to diameter ratios is investigated by considering granular material-structure interaction
using ABAQUS finite element package. Silo wall is modeled by shell elements and its behavior is
considered elastic, seismic behavior of granular material inside silo is highly nonlinear and requires a
complex nonlinear description of the granular material. The hypoplasticity theory describes the stress
rate as a function of stress, strain rate and void ratio. The granular material is modeled by solid elements
and its behavior is considered with a hypoplastic constitutive model, for modeling of interaction
between silo wall and granular material, surface to surface contact with coulomb friction law is
considered between silo wall and granular material. The results show that the seismic behavior of silos is
dependent on the height to diameter ratio of the silo. While considering a constant value for the
distribution of acceleration in the height of silo leads to conservative design pressures for a squat silo
based on Eurocode 8, this assumption is not conservative for a slender silo.

S
f

o
e

v
i
h

Keywords Steel silo; Seismic behavior; Finite element method; Hypoplasticity; Granular materialstructure interaction; Surface to surface contact.

. .

. ABAQUS
.
.
.

.
.


. Eurocode 8

c
r

1. INTRODUCTION
Silos are structures which are used for storing
granular materials like grain, coal and other
granular materials. Silos should be designed
against earthquake in earthquake-prone areas.
During earthquake silo wall experiences additional
IJE Transactions B: Applications

stresses resulting from unsymmetrical pressure

distributions in the silo. In addition the dynamic
loads lead to compaction of granular material
inside silo. In silo design based on ACI 313 [1]
wall pressures from such effects are not taken into
account. The system is reduced to a cantilever
beam with several point masses being situated on
Vol. 25, No. 1, February 2012 - 27

www.SID.ir

top of each other to calculate appropriate

additional static horizontal loads. About 80 percent of
actual weight of stored material should be
considered as effective weight for calculating
masses. Also 80 percent effective weight in ACI
313 resulted from agreement among Committee
313 members who wrote the original standard but
had not been verified by experiment at that time.
Subsequently Harris and von Nad [2] performed
shaking tests on relatively rigid steel model silos
filled with sand and wheat. The effective weight
coefficients from their experiments confirmed that
80 percent value considered in ACI 313. Eurocode
8 part 4 [3] considers additional horizontal
pressures resulting from earthquake effects with
simple relations. Eurocode 8 Part 4 has proposed
that if more accurate evaluations are not
undertaken, the global seismic response and the
seismic action effects in the supporting structure
may be calculated assuming that the particulate
contents of the silo move together with the silo
shell and modeling them with their effective mass,
the contents of the silo may be taken to have an
effective mass equal to 80 percent of their total
mass.
There are few researches that have tried to
investigate the behavior of silos under earthquake
loading with considering granular materialstructure interaction. Braun and Ebil [4] are the
first researchers that have proposed hypoplasticity
for modeling of granular material inside silos
under earthquake loading. They have used a simple
hypoplastic model. The behavior of granular
material is incrementally nonlinear even at low
strains. The hypoplasticity theory describes the
stress rate as a function of stress, strain rate and
void ratio. It can model the nonlinear and inelastic
behavior of granular material. Holler and
Meskouris [5] have tried to investigate the
behavior of silos under earthquake loading. They
have tested different hypoplastic models and
concluded that von Wolffersdorfs hypoplastic
model [6] with intergranular strain extension [7] is
the most effective material law for describing the
time dependent cyclic behavior of granular
material. The results of their research show that the
provisions given in the Eurocode yield good results
for the slender silos. While for squat silos the
results are too conservative, the material
parameters presented for granular material inside
silo in their research lead to high values of stiffness
for granular material inside silo. In this paper the

behavior of three steel silos with different height to

diameter ratios considering granular materialstructure interaction is investigated under
earthquake excitation using von Wolffersdorfs
hypoplastic model with intergranular strain
extension. Granular material inside silo models is
considered to be a type of sand which parameters
are available in technical literature.

2. PRESSURES UNDER EARTHQUAKE

LOADING IN EUROCODE 8 PART 4
In Eurocode 8 part 4, it is mentioned that if the
mechanical properties and the dynamic response of
the particulate solid are not explicitly and
accurately counted for the analysis, the
earthquake effect should be represented through an
additional normal pressure on the silo wall. In
circular silos, the additional normal pressure on the
wall may be taken as the following equation:

D
I

S
f

ph, s = ph, so Cos

(1)

v
i
h

o
e

where ph, so is the reference pressure and is

the angle between the radial line to the point of
interest on the wall and the direction of the
horizontal component of the seismic action. The
distribution of ph, s in the section of silo is
shown in Figure 1(a).

c
r

28 - Vol. 25, No. 1, February 2012

ph, s

ph, so

(a)

(b)

Figure 1. (a) The distribution of ph, s in the section

of silo, (b) The distribution of ph, so in the height of
silo

At points on the silo wall at a vertical distance x

from a flat bottom or the apex of a conical or
pyramidal hopper, the reference pressure ph, so
may be taken as:
IJE Transactions B: Applications

www.SID.ir

ph, so = ( z ) min(rs *, 3 x)

this paper von Wolffersdorffs hypoplastic

constitutive model with intergranular strain
extension is used for modeling of granular material
inside silo.
The general stress-strain relation in the hypoplastic
model with intergranular strain concept is:

(2)

(z ) is the ratio of the response acceleration of

the silo at a vertical distance z from the equivalent
surface of the stored contents, to the acceleration
of gravity. is the bulk unit weight of the
particulate material in the seismic design situation.
rs * is defined as below:
rs * = min(hb , d c / 2)

T= M :D

(4)

(3)

hb is the overall height of the silo, from a flat

bottom or the hopper outlet to the equivalent
surface of the stored contents and d c is the inside
dimension of the silo parallel to the horizontal
component of the seismic action. The distribution
of ph, so in the height of silo by considering
(z ) equal to a constant value (The ratio of
response acceleration at the center of gravity of the
particulate material to the acceleration of gravity)
is shown in Figure 1(b).

T is objective Jaumann stress rate, D is stretching

rate and M is a fourth-order tensor that represents
stiffness. The intergranular strain is obtained by
accumulation of D t and is a second-order
tensor. is the normalized magnitude of .

D
I

S
f

is the Euclidean norm of a tensor (e.g.

= ij ij ).
^

v
i
h

c
r

IJE Transactions B: Applications

is the direction of intergranular strain and is

o
e

3. HYPOPLASTICITY THEORY FOR

MODELING OF GRANULAR
MATERIAL
Hypoplasticity is a class of incrementally nonlinear
constitutive models that are developed to predict
the behavior of soils. The basic structure of the
hypoplastic models has been developed during
1990's at the University of Karlsruhe.
Hypoplasticity is a framework for the description
of mechanical behavior of granular materials. The
hypoplastic material laws describe the stress rate as
a function of stress, strain rate and void ratio and
are well for modeling of cohesionless granular
materials. Hypoplasticity can model the nonlinear
and inelastic behavior of soils due to its rate-type
formulation that ensures a realistic modeling of
loading and unloading paths. von Wolffersdorffs
hypoplastic constitutive model [6] can model the
nonlinear behavior of granular materials very well
but it has some drawbacks for application to cyclic
loadings. The most significant shortcoming of this
model is an excessive accumulation of
deformations for small stress cycles that is called
ratcheting. To solve this significant shortcoming,
Niemunis and Herle [7] presented an extension for
von Wolffersdorffs hypoplastic constitutive model
by introducing the intergranular strain concept. In

(5)

defined as below:
^
/
=
0

for 0

(6)

for = 0

Material stiffness can be calculated from the

following equation:

M = [ mT + (1 ) mR ]L +
^ ^
^
^

(1 mT )L : + N for : D > 0

^ ^
^
(mR mT )L :
for : D 0

(7)

The evolution equation for the intergranular strain

tensor is:
^ ^

(I r ) : D
=
D

for : D > 0
^

(8)

for : D 0

L is a fourth-order tensor and N is a secondorder tensor. The constitutive tensors L and N

are functions of stress and void ratio that are
defined in Equations 9 and 10.
Vol. 25, No. 1, February 2012 - 29

www.SID.ir

L = fb f e

^ ^

1
^

( F 2 I + a 2 T T)

fb =

(9)

T:T

N = f d fb f e

Fa
^

(10)

T:T

(11)

(12)

trT n
ei
ec
ed
= = = exp

ei 0 ec 0 ed 0
hs

D
I

1 is the unit tensor of second-order.

a=

3 (3 sinc )
2 2 sinc

F=

1
2 tan
1
tan 2 +

tan
8
2 + 2 tan cos 3 2 2

(14)

(15)

v
i
h
(16)

3/2

[tr( T* )]

In the above mentioned equations tensors of

second-order are denoted with bold letters and
tensors of fourth-order with calligraphic letters, in
addition different kinds of tensorial multiplication

c
r

^ ^

are used (e.g. M : D = Mijkl Dkl , = ij kl ,

^

N = N ij kl , : D = ij Dij ).

f e and f d are pycnotropy functions.

e
fe = c
e

e ed
f d =
ec ed

(17)

fb is barotropy function.
30 - Vol. 25, No. 1, February 2012

S
f

(18)

ed 0 is conventional minimum void ratio and ei 0 is

maximum possible void ratio at zero pressure. hs
is granular hardness that is a pressure-independent
stiffness and n is an exponent, appearing in the
power law for proportional compression. and
are exponents to be calculated from the triaxial
peak friction angle. Five additional material
parameters are required for intergranular strain
extension. R, mR , mT , r and are
intergranular strain parameters. The parameter R is
the maximum intergranular strain. The maximum
value of intergranular strain can be found from
stress-strain curves obtained either from so-called
dynamic tests or from static tests with strain
reversals. The incremental stiffness remains
approximately constant within a certain strain
range. The size of this range can be identified with
the constant R. Factors mR and mT are the
increase factors of stiffness for each load reversal
in the 180 degrees and 90 degrees directions
compared to the stiffness in the 0 degrees
direction. In order to determine the constants mR
and mT comparative tests at fixed values of T , e
and D but with different should be performed.
The parameters and r are used for smoothing
of stiffness change. The parameter can be

o
e

tan = 3 T*

cos 3 = 6

(20)

angle, ec 0 is the conventional maximum void ratio,

(19)

von Wolffersdorffs hypoplastic model requires

eight material parameters. c is critical friction

(13)

tr( T*3 )

minimum density), ec (critical void ratio) and ed

(maximum density) decrease with mean pressure
according to the following equation:

T is the stress tensor.

^
^ 1
T* = T- 1
3

1 n

These functions take into account the influence of

density and mean pressure. Three characteristic
void ratios ei (during isotropic compression at the

I is the unit tensor of fourth-order.

T = T / trT

1 + ei ei 0 trT

ei ec 0 hs

e e
3 + a 2 a 3 i 0 d 0

ec 0 ed 0

(T+ T*)

hs
n

IJE Transactions B: Applications

www.SID.ir

TABLE 1. Dimensions of the silo models

calibrated from cyclic test with small strain

amplitudes. The parameter r influences the
evolution of intergranular strain.

4. MODELING

Model

In this research three steel silos with different

height to diameter ratios were considered.
Hochstetten sand [8] was considered as granular
material inside silos. The mass density of sand was
considered equal to 1500 kg/m3. Dimensions of
silo models are presented in Table 1. ABAQUS
finite element package [9] was used for finite
element modeling. von Wolffersdorffs hypoplastic
constitutive model with intergranular strain
extension implemented in the form of UMAT [10]
for ABAQUS was used for modeling of granular
material inside silos. 8-noded solid element C3D8
was used for modeling of granular material inside
silos. The parameters of hypoplastic model for
sand are presented in Tables 2 and 3.
The initial value of void ratio considered for
granular material inside silos was 0.7. 4-noded
shell element S4 was used for modeling of silo
wall and silo bottom. The Modulus of elasticity of
steel wall of silos was considered equal to 2105
MPa. For decreasing the computation time only
half of silo was modeled and symmetric boundary
conditions were used at the center of silo and
granular material. The finite element mesh of silo
models and granular material inside silos is shown
in Figure 2. The interface between silo wall, silo
bottom and the granular material inside silo was
modeled by the contact pair algorithm provided
in ABAQUS, ABAQUS standard uses pure
master-slave contact. In pure master-slave contact,
one of the two surfaces comprising a contact pair is
assigned as the master surface and the other
surface as the slave surface. The surfaces on the
silo wall and silo bottom were considered as
master surface and the external surfaces on the
granular material that are in contact with silo wall
and silo bottom were considered as slave surface.
Coulombs friction law was used for modeling of
friction. The friction coefficient was set to be 0.4.
For the contact constraint, the penalty contact
algorithm was considered, which is similar to
introducing stiff springs between the two surfaces
to prevent them from penetration.

IJE Transactions B: Applications

Silo wall thickness t (m)

Model 1

10

10

0.01

Model 2

20

10

0.03

Model 3

30

0.05

TABLE 2. The parameters of von Wolfersdorffs

hypoplastic model

Granular
material
Hochstetten
sand

hs

( )

(N/m2)

33

1500106

ed 0

ec 0

ei 0

0.28

0.55

0.95

1.05

0.25

1.5

D
I

TABLE 3. Additional parameters for intergranular

strain concept
Granular material

Hochstetten sand

0.0001

S
f

o
e

v
i
h

c
r

Silo height H (m) Internal diameter D (m)

mR

mT

0.5

Model 1

Model 2

Model 3

Figure 2. The Finite element mesh of silo models and

granular material

5. ANALYTICAL PROCEDURE
The analysis includes two steps. The first step is
applying gravity loads, which were applied
statically. After applying gravity loads, earthquake
excitation was applied to the silo in the second
step. For applying of earthquake acceleration to the
Vol. 25, No. 1, February 2012 - 31

www.SID.ir

silo implicit dynamic analysis was used. The

earthquake acceleration applied to the silo models
is shown in Figure 3. The earthquake acceleration
was generated by SeismoMatch software [11] to be
approximately compatible with the spectrum of
Eurocode 8 [12] for soil Type B and design ground
acceleration of 0.2g. The response spectrum of
earthquake excitation is plotted in Figure 4.
Rayleigh damping was used for modeling of
viscous damping in silo structure. The value of
damping ratio was considered to be 0.05 in T and
0.33T, where T is the period of the first
translational mode of silo. For determination of
first mode period of silo models considered in this
paper, 80 percent of granular material mass was
applied to the silo wall uniformly. The period was
computed by eigenvalue analysis. The computed
values of T for silo models are presented in Table
4.

6. INVESTIGATING THE SILO

RESPONSE IN FREQUENCY DOMAIN

D
I

0.50
0.40
0.30
Acceleration (g)

0.20

0.00
-0.10
-0.20
-0.30
-0.40
0

v
i
h

Time (Sec)

10

Figure 3. Earthquake acceleration applied to the silo

models

c
r

0.7
0.6

Sa (g)

0.5
0.4
0.3
0.2
0.1
0
0

0.5

A
1

1.5

S
f

o
e

0.10

-0.50

After applying of earthquake acceleration to the

silo models, the velocity response of a point at the
top of silo in direction of applying earthquake
excitation is obtained in frequency domain by FFT
(Fast Fourier Transform). The FFT amplitudes of
velocity for the silo models are plotted in Figures
5-7. As shown in Figure 5, the frequency
corresponding to the highest peak of velocity
amplitude in model 1 is 4.15 Hz. There is another
peak with frequency of 3.723 Hz, while the
computed frequency value by applying 80 percent
of granular material mass to the silo wall is 7.394
Hz. The velocity response of model 2 in frequency
domain is plotted in Figure 6. As shown in this
figure, the frequency corresponding to the highest
peak of velocity amplitude in model 2 is 3.662 Hz.
However, the computed frequency value by
applying 80 percent of granular material mass to
the silo wall is 4.9 Hz. The velocity response of
model 3 in frequency domain is plotted in Figure 7.
As shown in this figure, the frequency
corresponding to the highest peak of velocity
amplitude in model 3 is 2.319 Hz, while the
computed frequency value by applying 80 percent
of granular material mass to the silo wall is 2.637
Hz. The results show that the dominant frequency
of models with height to diameter ratios of 1 and 2
has much difference with the frequency of first
translational mode by considering of 80 percent of
granular material mass as effective mass.
Nevertheless, in the silo with height to diameter
ratio equal to 5 these frequencies have smaller
difference.

0.045

Figure 4. Response spectrum of earthquake excitation

generated by SeismoMatch

TABLE 4. Computed values of period and frequency

by considering 80 percent of granular material mass as
effective mass
Model
Model 1
Model 2
Model 3

T (Sec)
0.135
0.204
0.379

32 - Vol. 25, No. 1, February 2012

F (Hz)
7.394
4.9
2.637

Velocity Amplitude (m/s)

Period (Sec)

0.04
0.035
0.03
0.025
0.02
0.015
0.01
0.005
0
0

10

Frequency (Hz)

Figure 5. Velocity response of a point at the top of silo

in frequency domain in model 1

IJE Transactions B: Applications

www.SID.ir

stiffness of granular material increases, for this

reason the assumed values of modulus of elasticity
for granular material increase from top layer to the
bottom layer. The values of modulus of elasticity
considered for different layers are presented in
Table 5. For determination of frequency it was
assumed that no separation can occur between silo
wall and granular material inside silo during
eigenvalue analysis. The frequency of first
translational mode for each silo model computed
by assuming 4 layers of granular material with
different values of stiffness is presented in Table 6
and is compared with dominant frequency value
obtained from applying earthquake acceleration to
the silo model and frequency of first translational
mode obtained from eigenvalue analysis by
considering 80 percent of granular material mass
as effective mass.

Velocity Amplitude (m/s)

0.18
0.16
0.14
0.12
0.1
0.08
0.06
0.04
0.02
0
0

10

Frequency (Hz)

Figure 6. Velocity response of a point at the top of silo

in frequency domain in model 2

Velocity Amplitude (m/s)

1.2
1
0.8

D
I

0.6
0.4
0.2
0
0

S
f

10

TABLE 5. Assumed values of modulus of elasticity for

each layer of granular material

Frequency (Hz)

Figure 7. Velocity response of a point at the top of silo

in frequency domain in model 3

Modulus of elasticity (MPa)

o
e

Layer 1

7. DETERMINATION OF FREQUENCY
BY CONSIDERING ELASTIC
BEHAVIOR FOR GRANULAR
MATERIAL

v
i
h

In fact by using von Wolffersdorffs hypoplastic

model with intergranular strain extension the
behavior of granular material inside silo is
incrementally nonlinear and the stiffness of
granular material in each integration point changes
during earthquake excitation. For understanding
the reason of difference between silo frequency
computed by considering 80 percent of granular
material mass as effective mass and frequency
obtained from the silo response under earthquake
excitation, the granular material inside silo was
considered to behave elastically and eigenvalue
analysis was performed for computation of silo
frequency. In each silo model granular material
was divided into four layers with different values
of modulus of elasticity. The values of modulus of
elasticity for these layers were determined by
judgment based on mean of three values of
tangential stiffness components in three directions.
By moving from top to the bottom of silo due to
higher pressures existing in granular material the

c
r

IJE Transactions B: Applications

Model 1

Model 2

Model 3

15

20

30

Layer 2

40

80

90

Layer 3

90

140

130

Layer 4

130

180

150

The deformed shapes of silo and granular material

inside silo in the first translational mode computed
by considering four layers of granular material
with elastic behavior are presented in Figure 8 for
all models. As shown in Figure 8, in model 1 that
has the lowest value of height to diameter ratio, the
largest displacements in the first mode have
occurred in the first layer of granular material near
the top surface of granular material. In model 2
with height to diameter ratio equal to 2 still the
largest displacements in the first mode have
occurred in the first layer of granular material near
the top surface of granular material. But, the
difference between largest displacements in silo
wall and first layer of granular material is less than
the corresponding difference in model 1. In model
3, with height to diameter ratio equal to 5, the
difference between largest displacements in silo
wall and first layer of granular material is less than
other models. It seems that in models with lower
Vol. 25, No. 1, February 2012 - 33

www.SID.ir

height to diameter ratios due to higher stiffness of

silo structure the dominant frequency obtained
from FFT of the response is dependent on the
stiffness of granular material. In models 1 and 2
that the silo structure is stiffer, the stiffness of
granular material in parts near the top surface of
granular material has a significant participation in
the dominant frequency of response. As shown in
Table 6, the computed values of frequency by
considering four layers of granular material with
elastic behavior in models 1 and 2 still have much
difference with the values of dominant frequency
obtained from FFT. The reason is that the value of
modulus of elasticity assigned to the first layer of
granular material overestimates the stiffness value
for the part of granular material which is situated
near the surface of granular material inside silo. In
model 3, the value of dominant frequency obtained
from FFT of the response has good correlation
with the value of frequency obtained from
eigenvalue analysis by considering four layers of
granular material with elastic behavior. The reason
is that in model 3 which is a slender silo due to
flexibility of silo structure, the stiffness of granular
material does not have a significant participation in
the dominant frequency of response.

Model
Model 1
Model 2
Model 3

F (Elastic Material)
(Hz)
6.056
4.343
2.431

F (FFT) F (80 Percent)

(Hz)
(Hz)
4.15
7.394
3.662
4.9
2.319
2.637

c
r

8. ENVELOPES OF DYNAMIC
PRESSURE

The envelopes of dynamic pressure in direction of

earthquake excitation for right and left sides of
silos are plotted versus height in Figures 9-11 for
all models. For obtaining the envelopes of dynamic
pressure, at first the time history of dynamic
pressure at each node is calculated, for this purpose
the time history of contact pressure is subtracted
from the value of contact pressure at the end of
gravity step. Then, the maximum value of dynamic
pressure at each node is considered as the envelope
value of dynamic pressure. The envelopes of
dynamic pressure are compared with the pressure
distribution proposed by Eurocode 8 part 4 [3]. For
34 - Vol. 25, No. 1, February 2012

D
I

S
f

o
e

v
i
h

TABLE 6. The values of frequency computed by

different methods

calculation of Eurocode pressure distribution ( z )

is considered constant. ( z ) is obtained from 5
percent damped spectrum of Eurocode 8 [12] for
soil type B and design ground acceleration of 0.2g
with assuming period computed by considering 80
percent of granular material mass as effective
mass. Figure 9 shows the envelopes of dynamic
pressure in model 1. As illustrated in this figure,
except in the lowest part of silo wall, the envelope
values of dynamic pressure are lower than the
proposed pressure by Eurocode. Therefore, it can
be concluded that assuming a constant value for
( z ) in squat silos is a rational assumption. Figure
10 shows the envelopes of dynamic pressure in
model 2. As illustrated in this figure, in addition to
the lowest part of silo wall in few points at the
upper half of silo height, the envelope values of
dynamic pressure have exceeded the proposed
pressure by Eurocode. Figure 11 shows the
envelopes of dynamic pressure in model 3. As
illustrated in this figure, the envelope values of
dynamic pressure have considerably exceeded the
proposed pressure by Eurocode at the upper half of
silo height. Therefore, it can be concluded that due
to impact of granular material into the silo wall
assuming a constant value for ( z ) in a slender
silo is not a conservative assumption.

9. CONCLUSIONS
In slender silo with height to diameter ratio equal
to 5, the values of frequency computed by all
methods are near each other. However, in silos
with lower height to diameter ratios the differences
between frequencies computed by different
methods are significant. It can be concluded that in
silos with lower height to diameter ratios, the
vibration of granular material inside silo in parts
situated near the top surface of granular material
plays an important role in the dominant frequency
of response. The stiffness of granular material
situated in layers near the top surface of granular
material controls the dominant frequency of
response.
In model with height to diameter ratio equal to
1, the Eurocode pressure distribution calculated
assuming a constant value for (z ) is
approximately conservative. By increasing this ratio
IJE Transactions B: Applications

www.SID.ir

Model 1

Silo

Granular material

S
f

Model 2

c
r

Silo

o
e

v
i
h

Silo

D
I

Granular material

Model 3

Granular material

Figure 8. The deformed shapes of silo and granular material inside silo in the first translational mode computed by
eigenvalue analysis considering four layers of granular material with elastic behavior

IJE Transactions B: Applications

Vol. 25, No. 1, February 2012 - 35

www.SID.ir

10

10

9
8

Dynamic Pressure
Eurocode 8 Pressure

Eurocode 8 Pressure

Height (m)

Height (m)

Dynamic Pressure

5
4

5
4

1
0

0
0

10

20

30

40

50

10

20

30

40

50

Pressure (KPa)

Pressure (KPa)

Left Side

Right Side

Figure 9. The distribution of dynamic pressure envelopes versus height in direction of earthquake excitation for right
and left sides of silo in model 1
20

20

18

18

S
f

16

Dynamic Pressure
Eurocode 8 Pressure

Dynamic Pressure

Eurocode 8 Pressure

14

14

12

12

Height (m)

Height (m)

16

D
I

o
e

10
8

10
8
6

iv

0
0

10

20

30

40

50

60

Pressure (KPa)

Left Side

h
c

10

20

30

40

50

60

Pressure (KPa)

Right Side

Figure 10. The distribution of dynamic pressure envelopes versus height in direction of earthquake excitation for right
and left sides of silo in model 2

r
A

30

25

20

Dynamic Pressure
Eurocode 8 Pressure

15

25

Height (m)

Height (m)

20

30

Dynamic Pressure
Eurocode 8 Pressure

15

10

10

0
0

10

20

30

40

50

60

Pressure (KPa)

Left Side

70

80

10

20

30

40

50

60

70

80

Pressure (KPa)

Right Side

Figure 11. The distribution of dynamic pressure envelopes versus height in direction of earthquake excitation for right
and left sides of silo in model 3

36 - Vol. 25, No. 1, February 2012

IJE Transactions B: Applications

www.SID.ir

in model with height to diameter ratio equal to 2,

the envelope values of dynamic pressure in few
points at the upper half of silo height have
marginally exceeded the Eurocode pressure
calculated assuming a constant value for ( z ) .
Nevertheless, when the height to diameter ratio
increases to 5, the envelope values of dynamic
pressure at the upper half of silo height have
considerably exceeded the Eurocode pressure
calculated assuming a constant value for ( z ) . It
can be due to impact of granular material into the
silo wall. Therefore, it can be concluded that
considering a constant value for ( z ) in a slender
silo is not a conservative assumption.

3.

4.

5.

6.

7.

8.

1.

2.

10. REFERENCES

9.

American Concrete Institute. ACI 313. Standard

practice for design and construction of concrete silos
and stacking tubes for storing granular materials,
(1997).
Harris, E.C. and von Nad, J.D., Experimental
determination of effective weight of stored material for

10.
11.
12.

use in seismic design of silos, ACI Journal, Vol. 82,

(1985), 828833.
European Committee for Standardization, Eurocode 8:
Design of structures for earthquake resistance, part 4:
silos, tanks and pipelines, (2006).
Braun, A. and Eibl, J., Silo pressures under earthquake
loading, X International Conference on Reinforced
and Post-tensioned Concrete Silos and Tanks.,
Cracow, Poland, (Nov. 23-25, 1995), 1995.
Holler, S. and Meskouris, K., Granular material silos
under dynamic Excitation: numerical simulation and
experimental validation, Journal of Structural
Engineering, Vol. 132, No. 10, (2006), 1573-1579.
vonWolffersdorff, P.A., A hypoplastic relation for
granular materials with a predefined limit state surface,
Mechanics of Cohesive Materials, Vol. 1, No. 3,
(1996), 251-271.
Niemunis, A. and Herle, I., Hypoplastic model for
cohesionless soils with elastic strain range, Mechanics
of Cohesive Materials, Vol. 2, No. 4, (1997), 279-299.
Masin,
D.,
PLAXIS
implementation
of
hypoplasticity, Plaxisbv, Delft, (2010).
DassaultSystmes, ABAQUS users manual, version
6.9. , (2009).
www.soilmodels.info.
www.seismosoft.com.
European Committee for Standardization. Eurocode 8:
Design of structures for earthquake resistance, part 1:
General rules, seismic actions and rules for buildings,
(2003).

D
I

S
f

o
e

v
i
h

c
r

IJE Transactions B: Applications

Vol. 25, No. 1, February 2012 - 37

www.SID.ir